Command Palette
Search for a command to run...
Open Source German Distant Speech Recognition: Corpus and Acoustic Model
{and Chris Biemann Max Mühlhäuser Stefan Radomski Evandro Gouvea Arvid Lange Benjamin Milde Stephan Radeck-Arneth}
Abstract
We present a new freely available corpus for German distant speech recognition and report speaker-independent word error rate (WER) results for two open source speech recognizers trained on this corpus. The corpus has been recorded in a controlled environment with three different microphones at a distance of one meter. It comprises 180 different speakers with a total of 36 hours of audio recordings. We show recognition results with the open source toolkit Kaldi (20.5% WER) and PocketSphinx (39.6% WER) and make a complete open source solution for German distant speech recognition possible.
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| speech-recognition-on-tuda | Kaldi | Test WER: 20.5% |
| speech-recognition-on-tuda | PocketSphinx | Test WER: 39.6% |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.